Staphylococcus aureus bacteremia, is a serious, common infection. The broad, long-term objectives of this proposal are to identify clinical and bacterial features that determine the clinical severity of S. aureus bacteremia, and to use these features to develop a clinically useful predictive model for assessing risk of individual patients with this infection. The Specific Aims of this proposal are 1.) To prospectively identify patients with S. aureus bacteremia at high risk for complications and death and 2.) To define the impact of selected bacterial characteristics (virulence factors) on the clinical outcome of patients with S. aureus bacteremia. This will be a prospective cohort study with outcomes data. The methods to be used include three related components. First, logistic regression will be employed on a large (n = 502) existing cohort of prospectively identified patients with S. aureus bacteremia to identify clinical features associated with complicated Staphylococcal infection. Second, assays for selected virulence factors will be performed on the stored S. aureus isolates from this large existing patient cohort. These assays will be performed through existing collaborations with established investigators. Results of these assays will then be correlated with patient outcomes. Third, prospective collection of clinical data and bacterial isolates from patients with S. aureus bacteremia will continue throughout the grant period in order to create a new cohort of patients with S. aureus bacteremia. This cohort will be used to validate the findings of this investigation. Accomplishing the aims of this proposal will contribute to improved clinical management of patients with S. aureus bacteremia, and will add to the understanding of S. aureus virulence factors.